Study on Bond Degradation of Rebars in Cracked Concrete ...kanakubo/syllmas2020.pdf1 ABSTRACT This study aims to quantify the bond strength of reinforcing bars in cracked concrete

Study on Bond Degradation of Rebars in Cracked

Concrete due to Bar Corrosion

SYLL AMADOU SAKHIR

(Master's Program in Engineering Mechanics and Energy )

Advised by Kanakubo Toshiyuki

Submitted to the Graduate School of

Systems and Information Engineering

in Partial Fulfillment of the Requirements

for the Degree of Master of Engineering

at the

University of Tsukuba

March 2020

1

ABSTRACT This study aims to quantify the bond strength of reinforcing bars in cracked concrete. It

focuses on the more fundamental effect of the cracking itself (in absence of corrosion products).

To achieve this aim, the simulation of the cracking of concrete due to corrosion of reinforcing

bars by using an aluminum pipe embedded in concrete and filled with an expansion agent was

proposed as a novel method. With the increase of the crack width over elapsed time from the

filling of the expansion agent, a target crack width was obtained. The splitting cracks play a

fundamental role in reducing the bond, therefore two cases were set: “Single-split type” when

the splitting cracks are parallel or along the rebar and “Side split-type” when the splitting cracks

are perpendicular to the rebar.

Following to tension test of the ribbed aluminum pipe, pull-out test was conducted on 28

specimens divided into two categories as described above. In this study, the induced crack width

was taken as the main variable to find its influence on bond degradation.

These experiments confirmed that the expansion agent filled pipe is a promising method that

allows focusing on the cracking itself. The tensile test also shows that the strain in the axis, top

and between the ribs tended to increase over the time after the filling of the expansion agent. In

addition, at the beginning of the loading, with the increasing of the axial strain, the yield

strength and modulus of elasticity of the pipe filled with an expansion agent also increase.

However, the elongation at failure tended to decrease. The tensile strength was not affected.

In pull-out test, all specimens experienced failure due to splitting. A group of specimens

failed by newly generated splitting crack despite existing induced crack and other specimens

failed by the opening of induced crack that the pull-out strength load reduces exponentially with

the increase of crack width and this demonstrates that the surface crack width can potentially

be a good indicator to evaluate the bond strength degradation. The research has also shown that

the decrease of the pull-out strength is more severe in 18MPa specimens than in 30 MPa

specimens. In addition, the deterioration of the bond due to the induced cracks was more severe

in a “Side-split type” than in “Single split-type”.

Keywords: Concrete crack width, Rebar corrosion, Crack width, Expansion agent, Bond test,

Deterioration of bond strength

2

Table of Contents Study on Bond Degradation of Rebars in Cracked Concrete due to Bar Corrosion ..

鉄筋腐食によりひび割れが生じたコンクリート中の鉄筋の付着劣化に関する研究

ABSTRACT ..................................................................................................................... 1

List of Figures .................................................................................................................. 4

List of Tables ................................................................................................................... 5

Chapter 1 Introduction ........................................................................................................ 6

1.1. Background ............................................................................................................ 6

1.2. Objectives of this study .......................................................................................... 8

1.3. Methodology of this study ..................................................................................... 8

1.4. Outline of the thesis ............................................................................................. 11

Chapter 2 Tensile properties of ribbed aluminum pipe filled with an expansion agent

12

2.1. Introduction .......................................................................................................... 12

2.2. Background of crack simulation by aluminum pipe filled with an expansion

agent 12

2.3. Experiment outline ............................................................................................... 13

2.3.1. Specimen Overview ........................................................................................... 13

2.3.2. Loading and Measurement ............................................................................... 15

2.4. Experiment results ............................................................................................... 16

2.4.1. Strain before loading due to the expansion agent .......................................... 16

2.4.2. Tensile test results ............................................................................................. 18

2.5. Conclusions ........................................................................................................... 21

Chapter 3 Bond Degradation of Rebars in Cracked Concrete due to BAR: Single

Splitting Case ...................................................................................................................... 22

3.1. Introduction .......................................................................................................... 22


3.2.1. Aluminum pipe with ribs .................................................................................. 22

3.2.2. Pull-out specimen .............................................................................................. 22

3.2.3. Crack simulation by EAFP............................................................................... 24

3.2.4. Loading and measurement ............................................................................... 25



3.3.2. Results of pull-out test ....................................................................................... 30

3.3.2.1. Failure mode ............................................................................................... 30

3.3.2.2. Pull-out load and slip .................................................................................. 31

3.4. Bond strength degradation .................................................................................. 33

3

3.4.1. Comparison of the results with the fib model code 2010 ............................... 33

3.4.2. Bond degradation and surface crack width relationship .............................. 35

3.5. Conclusions ........................................................................................................... 37

Chapter 4 Bond Degradation of Rebars in Cracked Concrete due to Rebar corrosion:

Side-Splitting Case .............................................................................................................. 38

4.1. Introduction .......................................................................................................... 38


4.2.1. Pull-out specimen and materials ...................................................................... 38


4.2.2. Loading and measurement ............................................................................... 42



4.3.1. Results of pull-out test ....................................................................................... 46

4.3.1.1. Failure mode ............................................................................................... 46

4.3.1.1. Pull-out load and slip .................................................................................. 49

4.3.1.2. Crack opening versus slip .......................................................................... 52

4.4. Bond strength degradation .................................................................................. 53

4.4.1. Comparison of the results with the fib model code 2010 and JCI 1998 ....... 53

4.4.1. Bond degradation and surface crack width relationship .............................. 54

4.5. Conclusions ........................................................................................................... 56

Chapter 5 Conclusions ....................................................................................................... 57

DEDICACES ................................................................................................................. 58

REFERENCES .............................................................................................................. 59

4

List of Figures Fig. 1.1 Mechanical interaction between concrete and deformed steel bar ........................ 6

Fig. 1.2 Sequences and consequences of corrosion on reinforcements[2] ........................... 7

Fig. 1.3 Case of single-splitting ................................................................................................ 9

Fig. 1.4 Case of side-splitting .................................................................................................. 9

Fig. 1.5 Schematic illustration of the methodology followed in the present study ........... 10

Fig. 2.1 Concrete cracking with expansion agent ................................................................ 12

Fig. 2.2 Specimen for cracking confirmation ....................................................................... 13

Fig. 2.3 Crack width over time .............................................................................................. 13

Fig. 2.4 Scheme of aluminum pipe with rib ......................................................................... 14

Fig. 2.5 Picture of aluminum pipe with rib .......................................................................... 14

Fig. 2.6 Position of strain gauges ........................................................................................... 15

Fig. 2.7 Loading setup ............................................................................................................ 16

Fig. 2.8 Strain and time relationship after filling the expansion agent ............................. 17

Fig. 2.9 Failure of the specimen ............................................................................................ 19

Fig. 2.10 Stress-strain relationship ....................................................................................... 20

Fig. 3.1 Aluminum pipe with ribs ......................................................................................... 22

Fig. 3.2 Specimen detail ......................................................................................................... 23

Fig. 3.3 Picture of the mold .................................................................................................... 23

Fig. 3.4 Filling of expansion agent ........................................................................................ 24

Fig. 3.5 Loading and measurement setup ............................................................................ 25

Fig. 3.6 Teflon sheet ................................................................................................................ 26

Fig. 3.7 Picture of specimen at loading ................................................................................. 26

Fig. 3.8 Cracking of concrete after filling the expansion agent .......................................... 28

Fig. 3.9 Distribution of crack width ...................................................................................... 29

Fig. 3.10 Splitting with crack opening and newly generated crack ................................... 30

Fig. 3.11 Splitting with crack opening only .......................................................................... 30

Fig. 3.12 Load-slippage relationship ..................................................................................... 32

Fig. 3.13 Maximum pull-out load vs crack width ................................................................ 33

Fig. 3.14 Result in comparison with fib Model Code 2010 ................................................. 34

Fig. 3.15 Result and prediction model .................................................................................. 35

Fig. 4.1 Specimen details ........................................................................................................ 39

Fig. 4.2 Specimen mold .......................................................................................................... 39

Fig. 4.3 Filling of the expansion agent .................................................................................. 41

Fig. 4.4 Loading and measurement setup ............................................................................ 42

Fig. 4.5 Picture of specimen at loading ................................................................................. 43

Fig. 4.6 Teflon sheet ................................................................................................................ 43

Fig. 4.7 Cracking of concrete after filling the expansion agent .......................................... 44

Fig. 4.8 Distribution of crack width ...................................................................................... 45

Fig. 4.9 Failure mode .............................................................................................................. 47

Fig. 4.10 Damage inside the bond part ................................................................................. 48

Fig. 4.11 Pull-out load versus slip at free-end for 18 MPa specimens .............................. 50

Fig. 4.12 Pull-out load versus slip at free-end for 30 MPa specimens .............................. 51

Fig. 4.13 Maximum pull-out load versus crack width before loading ............................... 52

Fig. 4.14 Crack opening versus slip ...................................................................................... 52

Fig. 4.15 Comparison with fib model 2010 and JCI 1998 ................................................... 53

Fig. 4.16 Result and prediction model .................................................................................. 54

Fig. 4.17 Proposed evaluation formula from test result ...................................................... 55

5

List of Tables Table 2.1 Dimension of aluminum pipe with rib ................................................................. 14

Table 2.2 List of specimens .................................................................................................... 14

Table 2.3 Strain before loading ............................................................................................. 16

Table 2.4 Tensile test results.................................................................................................. 19

Table 3.1 Concrete proportion of used concrete ................................................................. 23

Table 3.2 Concrete mechanical properties ........................................................................... 24

Table 3.3 Specimen list ........................................................................................................... 24

Table 3.4 Crack level .............................................................................................................. 29

Table 3.5 Maximum crack width .......................................................................................... 29

Table 3.6 Test result list ......................................................................................................... 31

Table 3.7 fib model code 2010 ............................................................................................... 34

Table 3.8 Regression analysis result ..................................................................................... 35

Table 3.9 Comparison of the calculated and experimental values .................................... 36

Table 4.1 Specimen list ........................................................................................................... 40

Table 4.2 Concrete mix proportion ...................................................................................... 40

Table 4.3 Concrete mechanical properties ........................................................................... 40

Table 4.4 Mechanical properties of reinforcement ............................................................. 40

Table 4.5 Maximum crack width .......................................................................................... 46

Table 4.6 Test result list ......................................................................................................... 48

Table 4.7 Regression analysis result ..................................................................................... 54

6

Chapter 1 Introduction

1.1. Background To guarantee the composite action of reinforced concrete members, the bond at the steel-

concrete interface is an important property of reinforced concrete. It allows the forces to be

transferred from the reinforcement to the surrounding concrete in a structure. The scientists who

have contributed to the knowledge of the many aspects of bonding agree that the bond stress

includes three components: chemical adhesion, friction, and mechanical interlock (see Fig. 1.1)

[1]. When the rebar is in tension, the concrete crushing and splitting occur due to mainly the

mechanical interlock. It is considered that the bond between reinforcement and concrete has a

large influence on the structural behavior of RC members. Therefore, to get satisfactory

performance of RC structures, it is necessary to assure adequate bond.

However, the deterioration of RC structures due to the corrosion of rebar is now becoming

a demanding problem.

Fig. 1.1 Mechanical interaction between concrete and deformed steel bar

In recent decades, lots of RC structures are threatened by corrosion worldwide, causing huge

direct and indirect costs annually. As shown in Fig. 1.2 [2], bond deterioration can negatively

affect the structural properties of RC structures. This was confirmed in many studies. To better

estimate the mechanical properties of corroded RC structures through theoretical analysis or

numerical modeling, it is essential to fully understand how corrosion affects bond.

Many studies investigated the degradation of bond strength due to the corrosion of

reinforcement which causes cracks along bars. Relevant bond tests were carried out by many

researchers, these studies have developed numerical models which are in good agreement with

test results [3] [4] [5].

However, even if these current models can give a good prediction, there are still some

limitations when they are put into practice. One of the most significant current discussions is

that these models mostly use the steel section loss or mass loss as a variable, which is very

difficult to measure in existing structures that are under services conditions. Regarding the

corrosion of rebars in reinforced concrete, the surface crack width provides the clearest visual

manifestation. It is a key parameter to solve this problem by directly correlating the bond

deterioration of the surface crack width because it can be easily measured. A rough correlation

between bond strength reduction and surface crack width is proposed in the fib Model Code

2010 [6]. On another hand, most studies use accelerated electrical corrosion techniques. The

combination of the effect of cracking and the formation of rust around the bar can lead to

difficulties in analyzing the processes at a fundamental level. Also, when concrete cracks

7

without corrosion, those results could be difficult to use.

Very little is currently known about the potential correlations between the bond and the

surface crack width. However, further investigations with this respect should be performed to

give more experimental evidence.

Fig. 1.2 Sequences and consequences of corrosion on reinforcements[2]

8

1.2. Objectives of this study

The major aim of this study is to increase the knowledge of the bond behavior in reinforced

concrete structures with cracks due to corrosion, as well as investigating the possible links

between visual inspection data (crack width) and bond degradation. To achieve this goal, there

are some specific objectives which relate to the different stages of the research project:

• Testing the tensile properties of ribbed aluminum pipe filled with an expansion agent

• Investigating the bond degradation with different crack width along the reinforcing bars.

• Investigating the bond behavior with different crack width occurring on the surrounding concrete

1.3. Methodology of this study

The scientific approach of this study has been primarily based on experiments and

corresponding observations. The simulation of the cracking of concrete due to corrosion of

reinforcing bars by using an aluminum pipe embedded in concrete and filled with an expansion

agent was used as a novel method to simulate reinforcing bars in which volume expands due to

corrosion. Throughout this thesis, the abbreviation ‘EAFP’ is used to refer to the expansion

agent filled pipe. With the increase of the crack width over elapsed time from filling of

expansion agent, a target crack width can easily be obtained.

First, for grasping the mechanical properties of EAFP, tensile tests of the aluminum pipes

with ribs are carried out. In addition, two series of experiments are carried out to study the effect

of induced crack width on bond degradation. The one end pull-out test is conducted with a short

bond length designed to enable the focus on local bond behavior. The splitting cracks plays a

fundamental role of reducing the bond, therefore two cases are set: “Single-split type” when the

induced cracks are along the rebar (Fig. 1.3) and “Side split-type” when the induced cracks are

located on the surroundings of concrete (Fig. 1.4). The simulated cracks on specimens are

measured before the tests. During the test, the pull-out load, slippage at the free end of the rebar

and the opening of the width of the induced crack are measured. These data are compared to

those in the existing literature. Regression analysis is also conducted to propose a simple model

for evaluating the bond degradation with crack width as the main variable. Fig. 1.5 shows a

schematic illustration of the methodology followed in this study.

9

Fig. 1.3 Case of single-splitting

Fig. 1.4 Case of side-splitting

10

Formulating the research question

Evaluation of experimental studies

Fig. 1.5 Schematic illustration of the methodology followed in the present study

How the cracking of concrete due to corrosion can affect the bond in

reinforced concrete structures?

Review of published

experimental studies

Literature study

(state-of-art)

Identification of

knowledge gaps

Definition of objectives

Design of

experiment

Tensile behavior of

ribbed EAFP

Expansion agent

Aluminum pipe with ribs

Filling of specimens

Induced cracks

Analytical models

Comparison with

previous research

Data analysis

Interpretation of

results

Regression analysis

New findings

Sim

ula

tion o

f co

ncr

ete

crac

kin

g d

ue

corr

osi

on

rebar

s

One

end P

ull

-out

exper

imen

ts

Pull-out load Slippage Induced

cracks

pening

Loading stages

Bond behavior

PHASE 1

PHASE 2

PHASE 3

Keywords

objectives Keywords

objectives

11

1.4. Outline of the thesis

The thesis consists of 5 chapters. In Chapter 1, the background, objectives, and methodology

of the study are provided. In order to grasp the mechanical properties of ribbed aluminum pipes

filled with an expansion agent when the pipe replaces reinforcing bars, the tensile tests of those

pipes are carried out, therefore Chapter 2 presents the tensile properties of ribbed aluminum

pipe filled with an expansion agent. Chapter 3 present the pull-out study of the Single-split

specimen when the splitting cracks are parallel or along the rebar. Chapter 4 outlines the pull-

out experience in case of “Side split”, in other words when the splitting cracks are

perpendicular to the rebar. Conclusions and suggestions for future research are placed in

Chapter 5.

12

Chapter 2 Tensile properties of ribbed aluminum pipe filled with an expansion agent

2.1. Introduction The possibility to simulate the cracking of concrete due to corrosion of reinforcing bars in

the laboratory in a relatively short time by using EAFP was demonstrated. In this chapter, in

order to grasp the mechanical properties of EAFP when the pipe replaces reinforcing bars,

tensile tests of those pipes are carried out.

2.2. Background of crack simulation by aluminum pipe filled with an expansion agent

Many studies have been conducted about the effects of reinforcing bar corrosion on the

structural performance of RC members using experiments that simulate cracks due to corrosion

expansion of reinforcing bars by an electrolytic corrosion test or by slits in the previous research,

but the conformity to real cracks is unclear. At the Kanakubo lab, in the University of Tsukuba,

an aluminum pipe embedded in concrete and filled with an expansion agent has been proposed

as a method to simulate reinforcing bars whose volume expands due to corrosion.

An expansion agent is mainly used for the destruction of rocks and RC structures. In powder

form, it expands when humified. Due to this expansion, cracks are generated in the concrete

(see Fig. 2.1).

Fig.2.2 shows the specimen used in order to confirm the possibility of crack simulation [7].

An aluminum pipe with 18 mm as outer diameter and 1 mm thickness was embedded in the

concrete. The compressive strength of the concrete was 25.3 MPa and the ratio of the water to

expansion agent was 30%. The specimen was placed as the axial direction of the pipe was set

vertically, and an expansion agent was filled from the top of the pipe

As can be seen in Fig.2.3, the possibility to simulate the cracking of concrete due to corrosion

of reinforcing bars in the laboratory in a relatively short time was confirm. Also, it was found

that the crack width can be easily targeted due to the fact that cracks width increases over time.

Fig. 2.1 Concrete cracking with expansion agent

13

Fig. 2.2 Specimen for cracking confirmation

Fig. 2.3 Crack width over time

2.3. Experiment outline

2.3.1.Specimen Overview

According to JIS G 3112 (Steel bars for concrete reinforcement), the ribs were machined on

an aluminum pipe. In order to simulate a deformed bar D19 with an aluminum pipe with an

outer diameter of 21.7mm and a thickness of 2.5mm was used. The details of the aluminum

pipe with ribs are shown in Fig.2.4 and Table 2.1. Table 2.2 summarizes the list of specimens.

The test was conducted on specimen filled with an expansion agent or not. For specimens

without expansion agent, three with ribs and three without ribs were set. 5 specimens with ribs

were filled with an expansion agent and a target strain before loading was taken as the main

parameter.

0 50 1000

0.1

0.2

0.3

Time(h)

Cra

ck

Wid

th(m

m) 18-15-3-Top

18-15-3-Center18-15-3-Bottom

14

Fig. 2.4 Scheme of aluminum pipe with rib

Fig. 2.5 Picture of aluminum pipe with rib

Table 2.1 Dimension of aluminum pipe with rib

Pipe diameter(d)

(mm)

Height of ribs (h)

(mm)

Interval between ribs(l)

(mm) h/l h/d l/d

21.70 1.50 13.40 0.11 0.07 0.61

Table 2.2 List of specimens

Specimen Ribs Expansion Agent

Target strain

before

loading

Number of specimens

P19-NRNF No No － 3

P19-RNF Yes － 3

P19-RF1 Yes 0.3% 1

P19-RF2 0.5% 1

P19-RF3 0.8% 1

P19-RF4 1.0% 1

P19-RF5 1.2% 1

15

2.3.2.Loading and Measurement

As shown in Fig.2.6, strain gauges were fixed, and the time course of axial strain after

expansion agent filling and circumferential strain on and between ribs were measured. The

variable factor was the target axial strain at the start of the tensile test, with a range of 0.3% to

1.2%.

Monotonic tensile loading was carried out using a universal testing machine of 500 kN

capacity. Fig.2.7 shows the loading and measurement method for tension tests. To prevent the

deformation of the grasped part by the chuck, round steel bars having a length of 80 mm were

inserted into the specimen ends. In addition to the load, the axial and circumferential strain were

measured.

With ribs

Without rib

Fig. 2.6 Position of strain gauges

16

Fig. 2.7 Loading setup

2.4. Experiment results

2.4.1.Strain before loading due to the expansion agent

Table 2.3 summarizes the strain due to the effect of the expansion agent before the loading.

For the specimen P19-RF5, due to the peeling of gauge during measurement after filling the

expansion agent, the circumferential strain was not measured. Fig.2.8 shows the progression of

each strain after filling the expansion agent. The temperatures in the figure were obtained from

the Japan Meteorological Agency located in Tsukuba city. The axial strain and circumferential

strain increased both over time, and the inter-rib circumferential strain was larger. The speed of

the reaction sharply increases between 4 to 10 hours after filling the expansion agent and tends

to slow down. 4 hours after filling the expansion agent, the strain evolution is greatly affected

by the temperature. This phenomenon is remarkably observed when the temperature exceeded

15 ° C.

Table 2.3 Strain before loading

Specimen Axial strain (%) Circumferential strain on top of

the rib (%)

Inter-ribs

Circumferential strain

(%)

P19-RF1 0.297 0.088 0.664

P19-RF2 0.521 0.328 1.13

P19-RF3 0.799 1.04 1.93

P19-RF4 0.892 1.21 2.54

P19-RF5 1.23 －* －*

* Due to the peeling of gauge during measurement after filling the expansion agent, these

strains were not measured.

17

Fig. 2.8 Strain and time relationship after filling the expansion agent

18

2.4.2.Tensile test results

The tensile test results are shown in Table 2.4. Fig.2.9 shows examples of the fracture of the

specimen after loading.

The tensile stress was calculated dividing the load by the cross-sectional area of the

aluminum pipe. For the specimen without ribs, the cross-section is 150.8 mm2 and 55.6 mm2

for one with ribs.

Fig.2.10 shows the stress-strain relationship of the loaded specimen. The yield strength of

specimens without expansion agent (P19-NRNF and P19-RN) was taken when the stress

becomes almost constant just after the elastic region. The yield strength of the other test

specimens was obtained by 0.2% offset strength from the axial strain.

The modulus of elasticity was calculated by the least-squares method in a section that can be

regarded as the elastic zone in the stress-strain curve. Similarly, the absolute value of Poisson’s

ratio was calculated using the inter-ribs strain. It was obtained by dividing the elastic modulus

by the lateral elastic modulus. Two gauges points were marked at the specimen before loading.

The percentage of elongation after failure was determined by dividing the deformation distance

by the initial distance before loading.

According to the result, there is no evidence that the presence of ribs and the filling of an

expansion agent has an influence on the tensile strength. For all specimens, the tensile strength

is from 210 to 220 MPa.

On another hand, for the specimen without an expansion agent, no significant differences in

yield strength and elastic modulus are found between the specimens with ribs and ones without

ribs.

At the beginning of the loading, with the increasing of the axial strain, the yield strength and

modulus of elasticity of the pipe filled with an expansion agent increase too. However, the

elongation at failure tended to decrease. The tensile strength was not affected.

It can be seen that the presence of the ribs tends to decrease the Poisson’s ratio. A possible

explanation for this might be that the occurrence of restraint in the circumferential direction at

the ribs decreases the transversal elastic modulus between ribs.

19

Table 2.4 Tensile test results

Specimen Max load

(kN)

Tensile

strength

(MPa)

Yield

strength

(MPa)

Young

modulus

(GPa)

Poisson’s ratio

(%)

Elongation

(%)

P19-NRNF 33.02 219 96.7 74.4 0.341 －

P19-RNF 12.00 216 102 78.2 0.247 8.5

P19-RF1 11.78 212 97.5 64.0 0.239 7.6

P19-RF2 11.67 210 121 63.6 0.226 7.0

P19-RF3 11.56 208 146 84.0 － 5.9

P19-RF4 11.84 213 158 70.2 0.162 5.3

P19-RF5 12.34 222 170 95.7 － 3.9

Fig. 2.9 Failure of the specimen

20

Fig. 2.10 Stress-strain relationship

1 2 3

100

200

0

STRAIN (%)

ST

RE

SS

(M

Pa)

P19-RF1P19-RF2P19-RF3P19-RF4P19-RF5

1 2 3

100

200

0STRAIN (%)

ST

RE

SS

(M

Pa

)

P19-RNF-1P19-RNF-2P19-RNF-3

1 2 3

100

200

0

STRAIN (%)

ST

RE

SS

(M

Pa)

P19-NRNF-1P19-NRNF-2P19-NRNF-3

21

2.5. Conclusions The aim of this chapter was to examine the mechanical properties of ribbed aluminum pipes

filled with an expansion agent when the pipe replaces reinforcing bars. The experiment

confirmed that:

The strain in the axis, top and between the ribs tended to increase over the time after the filling of the expansion agent.

There is no evidence that the presence of ribs and the filling of an expansion agent has an influence on the tensile strength.

The presence of the ribs tends to decrease the Poisson’s ratio At the beginning of the loading, with the increasing of the axial strain, the yield strength

and modulus of elasticity of the pipe filled with an expansion agent increase too.

However, the elongation at failure tended to decrease. The tensile strength is not

affected.

22

Chapter 3 Bond Degradation of Rebars in Cracked Concrete due to Rebar Corrosion: Single Splitting

Case

3.1. Introduction In this chapter, to develop a simple formula for the bond strength of a corroded bar with

surface crack width as a variable, the pull-out test is conducted on a concrete block with an

embedded bar following different corrosion crack width simulated by EAFP. The test is carried

out on a single split type specimen as described previously. One advantage of using this

simulation method is that it allows focussing on the effect of the cracking while ignoring the

section loss.

A rough correlation between bond strength reduction and surface crack width is suggested

by the fib Model Code 2010. In addition to that Japan Concrete Institute in the concrete structure

rehabilitation research committee report 1998 gave an evaluation formula of the bond

degradation as a function of corrosion, also an overview of the current research indicates that

existing studies have achieved a primary knowledge regarding the potential correlation between

bond and the surface crack width. However, there still a research gap in this respect.


3.2.1.Aluminum pipe with ribs

Fig. 3.1 shows an overview of a processed aluminum pipe with ribs set according to JIS G

3112. An aluminum pipe with 21.7 mm as outer diameter and 2.5mm thickness was used to

imitate the D19 rebar. The fundamental properties of the aluminum pipe and the expansion

agent filled pipe have been reported in Chapter 2.

Fig. 3.1 Aluminum pipe with ribs

3.2.2.Pull-out specimen

The pull-out specimen was designed as shown in Fig.3.2. The dimensions of the specimen

were 260×260×82mm and the aluminum pipe with ribs was embedded at 38mm from the

specimen side. The framework is shown in Fig. 3.3. The bond length of 51.6mm was chosen to

avoid the rupture of the pipe by tensile force. Moreover, an unbonded part was set to avoid cone

failure of concrete. The M6 coupler was fixed 100×100 mm at the up left position to set a LDVT

for measuring the slip of pipe. A π-type displacement transducer was placed on the concrete

cover to measure the opening of crack during loading. Table 3.1 shows the mix proportion of

used concrete and Table 3.2 shows the mechanical properties of concrete obtained from the

concrete cylinder test on the day of filling of expansion agent. Table 3.3 shows the list of

specimens. The test variable was set on the level of crack width induced by expansion agent

23

filled pipe before pullout loading. A total of 12 specimens were tested.

Fig. 3.2 Specimen detail

Fig. 3.3 Picture of the mold

Table 3.1 Concrete proportion of used concrete

Water-cement

ratio (%)

W/C

Cement

(kg/m3)

Water

(kg/m3)

Air Fine

Aggregate

(kg/m3)

Coarse

Aggregate

(kg/m3)

Water

Reducing

Agent

(kg/m3)

82.5 242 200 4.5 % 937 827 2.42

24

Table 3.2 Concrete mechanical properties

Compressive strength

(MPa)

Young’s modulus

(GPa)

Splitting tensile strength

(MPa)

17.8 22.8 1.78

Table 3.3 Specimen list

Specimen

name

Cover

Thickness

Expected

crack level

Number of

specimens

S-L1

38mm

Level 1

12 S-L2 Level 2

S-L3 Level3

3.2.3.Crack simulation by EAFP

The ratio of the water to an expansion agent was set to 30%. The specimen was placed as the

axial direction of the pipe was set vertically, and an expansion agent was filled from the top of

the pipe as shown in Fig.3.4. With the increase of the crack width over elapsed time after filling

of expansion agent, a target crack width was easily obtained by using the time after filling as a

parameter.

Fig. 3.4 Filling of expansion agent

Filling of expansion agent

25

3.2.4.Loading and measurement

The loading was carried out at Kanakubo lab in the University of Tsukuba. Fig. 3.5 shows

the general set-up for the pull-out test. The specimen was placed on the Teflon sheet and the

loading plate on which the hole with the same diameter corresponding to concrete cover in

order to not restrict the lateral deformation of concrete. This detail can be seen from Fig 3.6 and

Fig.3.7 The pipe was subjected to monotonic pull-out loading at a speed of 0.5mm/min with a

Universal Testing Machine. The measurement items are pull-out load, crack opening and

slippage of the pipe at the free end.

Fig. 3.5 Loading and measurement setup

26

Fig. 3.6 Teflon sheet

Fig. 3.7 Picture of specimen at loading

27



Fig. 3.8 shows an example of crack patterns after filling the expansion agent. The expansion

agent reaction was heavily influenced by the ambient temperature, so to control the width of

the crack, specimens were placed in variable temperature conditions. Also, because cracks

continued to grow after filling the expansion agent, crack width was measured only when

growth had stopped fully.

Cracks were categorized according to their width, as shown in Table 3.4. The maximum crack

width in every specimen is summarized in Table 3.5. To better observe the distribution of

cracking among the specimens, the maximum cracks widths are plotted in Fig 3.9.

28

(a) Top view

(b) Bottom view

(c) Side view

Fig. 3.8 Cracking of concrete after filling the expansion agent

29

Table 3.4 Crack level Table 3.5 Maximum crack width

Level Crack width

range

Specimen

name

Maximum crack

width (mm)

Level 1 ≤ 0.5mm S-1-L1 0.15

Level 2 0.5mm to 1.

0mm S-2-L1 0.1

Level 3 > 1. 0mm S-3-L1 0.2

Specimen name

explanation

S-4-L1 0.35

S-2-L2 0.5

S-1-L2 1.0

S-1-L3 1.5

S-2-L3 3.0

S-3-L3 2.0

S-4-L3 1.8

S-5-L3 2.0

S-6-L3 2.6

Fig. 3.9 Distribution of crack width

0

1

2

3

Specimen

Crack

wid

th (

mm

)

Cracked specimen

30

3.3.2.Results of pull-out test

3.3.2.1.Failure mode

All specimens experienced failure due to splitting. A group of specimens failed by newly

generated splitting crack despite existing longitudinal crack due to corrosion as shown in

Fig.3.10 and other specimens failed by widening of existing crack induced by the expansion

agent as can be seen in Fig.3.11. The results of pull-out tests are summarized in Table 3.6.

Before loading

After loading

Fig. 3.10 Splitting with crack opening and newly generated crack

Before loading

After loading

Fig. 3.11 Splitting with crack opening only

31

Table 3.6 Test result list

Specimen

name

Biggest

crack width

(mm)

At Maximum

Remarks Load

(kN)

Slippage

(mm)

S-1-L1 0.15 7.99 0.088 New crack

S-2-L1 0.1 8.39 0.146 Crack opening

S-3-L1 0.2 6.96 0.218 New crack

S-4-L1 0.35 5.79 0.144 New crack

S-2-L2 0.5 6.14 0.198 New crack

S-1-L2 1.0 6.99 0.890 New crack







3.3.2.2. Pull-out load and slip

The pull-out load versus slippage relationships are shown in Fig. 3.12. In the case of Level 3

cracks, the load – slippage relationship tends not to results in a severe decrease of load after

maximum pull out force. The maximum pull-out load versus crack width relationship is shown

in Fig. 3.13. It can be seen that the maximum load decreases as the crack width increases. As

expected, there is a significant correlation between residual bond strength and induced crack

width.

To better observe the degradation due to the induced crack, the maximum pull-out load is

normalized by calculated pull-out splitting strength as reported in the previous study [8] as a

non-cracked specimen using the following equation:

τb,max = 0.601⋅σt ⋅(ru/db) ⋅ cotα

where, τb,max : bond splitting strength, σt: splitting tensile strength of concrete, ru:C+db/2, db: diameter of the pipe (19mm), C: the thickness of cover concrete, α: the angle between the

longitudinal axis and splitting force (=34 degree).

32

Level 1

Level 2

Level 3

Fig. 3.12 Load-slippage relationship

0 1 2 30

5

10

Free-end slip(mm)

Load

(k

N)

S-1-L1(0.15mm)S-2-L1(0.10mm)Calculated value

0 1 2 30

5

10

Free-end slip(mm)

Load

(k

N)

S-3-L1(0.20mm)S-4-L1(0.35mm)Calculated value

0 1 2 30

2

4

6

8

10

Free-end slip (mm)

Load

(k

N)

S-1-L2(0.5mm)S-2-L2(1.0)Calculated valueNew Crack

0 1 2 30

2

4

6

8

10

Free-end slip(mm)

Load

(k

N)

S-1-L3(1.5mm)S-4-L3(1.8mm)S-5-L3(2.0mm)Calculated value

0 1 2 30

2

4

6

8

10

Free-end slip(mm)

Load

(k

N)

S-2-L3(3.0mm)S-3-L3(2.0mm)S-6-L3(2.6mm)Calculated value

33

Fig. 3.13 Maximum pull-out load vs crack width

3.4. Bond strength degradation

3.4.1.Comparison of the results with the fib model code 2010

In the fib Model Code 2010[6], the reduction in bond strength depending on the surface crack

width was introduced. For a certain range of surface crack width, the possible variation of the

pull-out load degradation can be predicted as shown in Table 3.7. The first development of

cracks at corrosion penetrations (i.e. reductions in rebar radius) of between 0.015mm and

0.040mm was reported. Also, a relationship between crack width and corrosion penetration as

corrosion progressed was derived.

In this study, the proposed relationship is compared with the test results in Fig. 3.14. The

prediction with Model Code 2010 is in good agreement with Level 1 (crack width ≤ 0.5mm)

and for the specimen with 2.6mm and 3mm as crack width as an extrapolation. However, for

Level 2 (crack width 0.5mm to 1.0mm), the fib Model Code 2010 gives underestimation. It

should be noted that the authors of the fib model code 2010 assume that the residual strength

of concrete structures is also affected by cross-section loss of steel. In this study, only on the

influence of the induced crack in the concrete cover is focused on. Only the bond degradation

of level 2 specimens was higher than the prediction from the fib Model Code. This finding was

unexpected and suggests that when the crack width is between 0.5mm to 1 mm, the influence

of the bar profile change becomes greater. Interestingly, Yang et al (2019) [9] found that

corrosion crack in concrete is concluded to be a more dominant factor than the corroded rebar

shape and rust accumulation in bond deterioration mechanism.

1 2 3

2

4

6

8

10

0Crack width (mm)

Ma

xim

um

lo

ad

(k

N)

34

Table 3.7 fib model code 2010

Corrosion

penetration

(mm)

Equivalent

surface crack

(mm)

Confinement

Residual capacity %

Bar type

Ribbed Plain

0.05 0.2-0.4

No links

50-70 70-90

0.1 0.4-0.8 40-50 50-60

0.25 1.0-2.0 25-40 30-40

0.05 0.2-0.4

Links

95-100 95-100

0.10 0.4-0.8 70-80 95-100

0.25 1.0-2.0 60-75 90-10

Fig. 3.14 Result in comparison with fib Model Code 2010

1 2 3

0.2

0.4

0.6

0.8

1

0Crack width (mm)

No

rm

ali

zed

lo

ad

Present studyModel Code 2010: upper limitModel Code 2010: lower limit

35

3.4.2.Bond degradation and surface crack width relationship

As mentioned in the literature review, the existing models related to bond deterioration focus

on the relationship between bond strength and mass loss, and they are mostly formulated as

linear, exponential or logarithmic equations. The present study proposes a simple formula for

the bond strength of corroded bar with surface crack width as a variable and thus contributes to

accurately assess the effects of deterioration.

As can be seen from the results of regression analysis in Table 3.8, an exponential equation

provides the best fit (higher R2). Therefore, the following equation can be used for the

prediction of the maximum pull-out load:

P(Wcr) = P0 ⋅ e-0.515Wcr

where P(Wcr): Pull-out strength in cracked concrete; P0: Pull-out strength of specimen

without crack; Wcr: Crack width.

Fig 3.15 shows the experimental result and the prediction formula comparison. The

calculated and experimental results are compared in Table 3.9. The ratios of the experimental

to calculated values are close to one, which indicates that the concrete crack width can

potentially be a good indicator to characterize bond strength degradation.

Table 3.8 Regression analysis result

Analysis Equations R2

Linear -0.306Wcr+1 0.66

Logarithmic 0.175ln(Wcr) + 0.4916 0.80

Exponential e-0.515 Wcr 0.91

Fig. 3.15 Result and prediction model

1 2 3

0.2

0.4

0.6

0.8

1

0Crack width (mm)

No

rm

ali

zed

load

Present studyPrediction model

36

Table 3.9 Comparison of the calculated and experimental values

Specimen

name

Biggest

crack

width

(mm)

Experimental

maximum

load

(kN)

Calculated

maximum

load

(kN)

Exp /Cal

S-1-L1 0.15 7.99 8.93 0.89

S-2-L1 0.1 8.39 9.16 0.91

S-3-L1 0.2 6.96 8.70 0.80

S-4-L1 0.35 5.80 8.06 0.72

S-2-L2 0.5 6.14 6.72 0.91

S-1-L2 1.0 6.99 7.46 0.94

S-1-L3 1.5 4.11 3.81 1.07

S-2-L3 3.0 2.12 2.05 1.03

S-3-L3 2.0 3.77 3.44 1.09

S-4-L3 1.8 4.69 5.20 0.90

S-5-L3 2.0 3.92 3.44 1.13

S-6-L3 2.6 2.30 2.52 0.90

37

3.5. Conclusions

In this chapter, the pull-out test is conducted on 12 concretes block with an embedded bar

following different corrosion crack width simulated by EAFP. The test is carried out on a single

split type.

The prediction with Model Code 2010 is in good agreement with Level 1 (crack width ≤

0.5mm) and for the specimen with 2.6mm and 3mm as crack width as an extrapolation.

There was a significant exponential correlation between pull-out strength and surface crack

width. Also, a simple formula to predict bond degradation is proposed by using the surface

crack width as the main variable.

38

Chapter 4 Bond Degradation of Rebars in Cracked Concrete due to Rebar Corrosion: Side-Splitting Case

4.1. Introduction

In this chapter, a concrete cracked by EAFP in bond test specimens has been designed to fail

with side splitting of the cover. Those specimens are subjected to a pull-out test and a

relationship between the maximum pull-out load of bar and surface crack width as a variable is

discussed.


4.2.1.Pull-out specimen and materials

The pull-out specimen was designed as shown in Fig.4.1. The dimensions of the specimen

were 260×170×170mm and the D19 rebar is embedded at 47.5mm (2.5db) from the specimen

cover side. The framework is shown in Fig. 4.2. A short bond length of 4 times the diameter of

the rebar equal to 76 mm was chosen to focus on the local bond. To simulate the cracking of

the surrounding concrete of the rebar, 2 aluminum pipes with a diameter of 22mm and a

thickness of 1mm were set to 50mm from the rebar. Moreover, an unbonded part was set to

avoid cone failure of concrete. The M6 coupler was fixed 100×50 mm at the upright position

to set a LDVT for measuring the slip of the free end of the bar. Two π-type displacement

transducers were placed on the top and two others on the east and west side of the specimen to

measure the opening of crack during loading. Table 4.1 shows the list of the specimens. Table

4.2 shows the mix proportion of used concrete and Table 4.3 shows the mechanical properties

of concrete obtained from the concrete cylinder test on the day of filling of expansion agent.

The reinforcement is a deformed bar of a nominal diameter of 19 mm. Its mechanical properties

are presented in Table 4.4.

39

Fig. 4.1 Specimen details

Fig. 4.2 Specimen mold

40

Table 4.1 Specimen list

Table 4.2 Concrete mix proportion

Concrete

target

strength

W/C

(%)

Unit weigth (kg/m3)

C W S G Ad

18MPa 78.5 245 192 942 852 2.45

30MPa 56.0 321 180 846 918 3.21

Table 4.3 Concrete mechanical properties

Concrete

target strength

Compressive

strength

(MPa)

Young modulus

(GPa)

Splitting strength

(MPa)

18MPa 21.8 19.0 2.26

30MPa 31.7 21.6 2.59

Table 4.4 Mechanical properties of reinforcement

Nominal

diameter

Tensile strength

(MPa)

Yield strength

(MPa)

Young modulus

(GPa)

19 535 366 193

Series Specimen

name

Concrete target

strength

Induced

crack

Number of

specimens

I

S.18.NC

18MPa

No 2

S.18.C.

Level 1 - 3

6

II

S.30.NC

30MPa

No 2

S.30. C

Level 1 - 3

6

41


The ratio of the water to the expansion agent was set to 30%. The specimen was placed as

the axial direction of the pipes was set vertically, and an expansion agent was filled from the

top of the pipes as shown in Fig.4 .3.

Fig. 4.3 Filling of the expansion agent

42

4.2.2.Loading and measurement

The loading was carried out at Kanakubo lab in the University of Tsukuba. Fig. 4.4 shows

the general set-up for the pull-out test. The specimen was placed on the Teflon sheet and the

loading plate on which the hole with the same diameter corresponding to concrete cover in

order to not restrict the lateral deformation of concrete. This detail can be seen from Fig 4.5 and

Fig.4.6. The D19 rebar was subjected to monotonic pull-out loading at a speed of 0.5mm/min

with a Universal Testing Machine. The measurement items are pull-out load, crack opening and

slippage of the D19 rebar at the free end.

Fig. 4.4 Loading and measurement setup

43

Fig. 4.5 Picture of specimen at loading

Fig. 4.6 Teflon sheet

44



Fig. 4.7 shows an example of crack patterns after filling the expansion agent. The expansion

agent reaction was heavily influenced by the ambient temperature, so to control the width of

the crack, specimens were placed in variable temperature conditions. Also, because cracks

continued to grow after filling the expansion agent, crack width was measured only when the

target crack width was attained.

The maximum crack width on the top side (bottom side at the filling) in every specimen is

summarized in Table 4.5. To better observe the distribution of cracking among the specimens,

the maximum cracks widths was are plotted in Fig 4.8.

Fig. 4.7 Cracking of concrete after filling the expansion agent

Top

de

West

side

45

Fig. 4.8 Distribution of crack width

0

1

2

Specimen

Crack

wid

th (

mm

)18MPa specimen30MPa specimen

46

Table 4.5 Maximum crack width

Specimen

name

Crack width (mm)

Top side East side West side

S.18.NC.1 0 0 0

S.18.NC.2 0 0 0

S.18.C.0.08 0.08 0.40 0.40

S.18.C.0.2 0.20 0.40 0.40

S.18.C.0.6 0.60 1.20 1.20

S.18.C.0.65 0.65 1.00 1.20

S.18.C.0.75 0.75 1.30 1.20

S.18.C.0.85 0.85 1.00 1.00

S.30.NC.1 0 0 0

S.30.NC.2 0 0 0

S.30.C.0.15 0.15 0.35 0.35

S.30.C.0.25 0.25 0.80 0.70

S.30.C.0.35 0.35 0.80 0.85

S.30.C.0.50 0.50 1.00 1.00

S.30.C.0.60 0.60 1.30 1.20

S.30.C.0.85 0.85 1.40 1.30

4.3.1.Results of pull-out test

4.3.1.1.Failure mode

All specimens experienced failure due to splitting. The specimens without crack failed by

single splitting due to the limited cover thickness. In the crack induced specimens, most of them

failed by side-splitting due to the opening of pre-existent cracks. In side-splitting failure, some

specimens presented newly side crack (named N1 with one new crack, named N2 with two new

cracks). All the failure modes are summarized in Fig 4.9. Depending on the failure mode,

concrete between the ribs was totally damaged or without any damage. This can be seen in Fig

4.10. The results of pull-out tests are shown in Table 4.6.

47

Fig. 4.9 Failure mode

BEFORE AFTER

BEFORE AFTER

AFTER BEFORE

BEFORE AFTER

SIDE SPLITTING N1

SIDE SPLITTING N2

SINGLE SPLITTING

SIDE SPLITTING

48

Fig. 4.10 Damage inside the bond part

Table 4.6 Test result list

Specimen name

Crack width

(mm)

At Maximum load

Remarks Load (kN) Slip (mm)

S.18.NC.1 0 50.51 0.444 Single split

S.18.NC.2 0 46.36 0.962 Single split

S.18.C.0.08 0.08 31.36 0.508 Side split N1

S.18.C.0.2 0.20 22.87 0.352 Side split N2

S.18.C.0.6 0.60 17.22 0.832 Side split O

S.18.C.0.65 0.65 25.10 0.121 Side split N1

S.18.C.0.75 0.75 4.19 1.102 Side split O

S.18.C.0.85 0.85 4.29 0.676 Side split O

S.30.NC.1 0 42.97 0.592 Single split

S.30.NC.2 0 61.73 0.166 Single split

S.30.C.0.15 0.15 49.46 0.348 Single split

S.30.C.0.25 0.25 31.25 0.308 Single split

S.30.C.0.35 0.35 26.86 0.468 Side split O

S.30.C.0.50 0.50 33.20 0.61 Side split O

S.30.C.0.60 0.60 19.77 0.488 Side split N1

S.30.C.0.85 0.85 12.53 0.724 Side split O

SINGLE SPLITTING SIDE SPLITTING

49

4.3.1.1.Pull-out load and slip

During the pull-out tests, the slippage of the D19 rebar is measured so the pull-out load versus

the slip relationship can be plotted in Fig. 4.11 for 18 MPa and in Fig.4.12 for 30MPa. A plot

of the pull-out load – slippage relationship of a typical uncracked and cracked specimen shows

that the slope of the initial ascending branch seems to not change with the present number,

indicating a no decrease in stiffness. As the pull-out load increase, the slip seems to increase in

an almost linear way. As soon as the bond strength is reached however, a steeper descending

curve is measured for uncracked specimens when compared to cracked ones. Hence a more

sudden bond degradation is noted due to the occurrence of new cracks or the opening of the

induced cracks.

The maximum pull-out load versus crack width before loading relationship is shown in Fig.

4.13. It can be seen that the maximum load decreases as the crack width increases. As expected,

there is a significant correlation between residual bond strength and induced crack width.

The decrease of the pull-out load is more severe in 18MPa specimens than in 30 MPa

specimens. A possible explanation for this might be that the 30MPa concrete has a stiffer

response than the 18MPa concrete in both compression and tension therefore the crack opening

may be delayed. Another possible explanation for this is that the confinement effect developed

by the same concrete cover is perhaps more effective in 30MPa concrete.

50

Fig. 4.11 Pull-out load versus slip at free-end for 18 MPa specimens

0 1 2 3 4 50

10

20

30

40

50

60

FREE-END SLIP (mm)

LO

AD

(k

N)

S.18.NC.1S.18.NC.2S.18.C.0.08

0 1 2 3 4 50

10

20

30

40

50

60

FREE-END SLIP (mm)

LO

AD

(k

N)

S.18.NC.1S.18.NC.2S.18.C.0.2

0 1 2 3 4 50

10

20

30

40

50

60

FREE-END SLIP (mm)

LO

AD

(k

N)

S.18.NC.1S.18.NC.2S.18.C.0.6

0 1 2 3 4 50

10

20

30

40

50

60

FREE-END SLIP (mm)

LO

AD

(k

N)

S.18.NC.1S.18.NC.2S.18.C.0.65

0 1 2 3 4 50

10

20

30

40

50

60

FREE-END SLIP (mm)

LO

AD

(k

N)

S.18.NC.1S.18.NC.2S.18.C.0.75

0 1 2 3 4 50

10

20

30

40

50

60

FREE-END SLIP (mm)

LO

AD

(k

N)

S.18.NC.1S.18.NC.2S.18.C.0.85

51

Fig. 4.12 Pull-out load versus slip at free-end for 30 MPa specimens

0 1 2 30

20

40

60

FREE-END SLIP (mm)

LO

AD

(k

N)

S.30.NC.1S.30.NC.2S.30.C.0.15

0 1 2 30

20

40

60

FREE-END SLIP (mm)

LO

AD

(k

N)

S.30.NC.1S.30.NC.2S.30.C.0.25

0 1 2 30

20

40

60

FREE-END SLIP (mm)

LO

AD

(k

N)

S.30.NC.1S.30.NC.2S.30.C.0.35

0 1 2 30

20

40

60

FREE-END SLIP (mm)

LO

AD

(k

N)

S.30.NC.1S.30.NC.2S.30.C.0.50

0 1 2 30

20

40

60

FREE-END SLIP (mm)

LO

AD

(k

N)

S.30.NC.1S.30.NC.2S.30.C.0.60

0 1 2 30

20

40

60

FREE-END SLIP (mm)

LO

AD

(k

N)

S.30.NC.1S.30.NC.2S.30.C.0.85

52

Fig. 4.13 Maximum pull-out load versus crack width before loading

4.3.1.2.Crack opening versus slip

The Crack opening versus slip is plotted in Fig.4.14. The crack opening is measured by

displacement transducers without including crack width before loading. In the case of plural

cracks, crack opening shows the summation of the width of cracks that occurred in the gauge

length of displacement transducers. With increasing of slip, the opening seems to start earlier

and develop faster in case of side-splitting without new crack. In the case of an apparition of

new cracks, the opening of the crack is fastest and reaches the highest value compared to other

specimens. However, in some specimens, no evidence was found.

Fig. 4.14 Crack opening versus slip

0.2 0.4 0.6 0.8 1

20

40

60

0Crack Width (mm)

Maxim

um

Load

(k

N)

18MPa Specimens30MPa Specimens

0 1 2 30

1

2

3

Free-end slip (mm)

Crack

op

en

ing (

mm

)

S.18.NC.1S.18.NC.2S.18.C.0.08S.18.C.0.2S.18.C.0.6S.18.C.0.65S.18.C.0.75S.18.C.0.85

0 1 2 30

1

2

3

Free-end slip (mm)

Cra

ck o

pen

ing (

mm

)

S.30.NC.1S.30.NC.2S.30.C.0.15S.30.C.0.25S.30.C.0.35S.30.C.0.50S.30.C.0.60S.30.C.0.85

53

4.4. Bond strength degradation

4.4.1.Comparison of the results with the fib model code 2010 and JCI 1998

The proposed relationship in the fib Model Code is compared with the test results in Fig.

4.15. The prediction with Model Code 2010 is in good agreement with S.30.C.0.25 and

S.30.C.0.30. However, for other specimens, our results are overestimated or underestimated

for 18 MPa specimens. It should be noted again that the authors of the fib model code 2010

assume that the residual strength of concrete structures is also affected by cross-section loss of

steel.

In addition to that, Japan Concrete Institute in the concrete structure rehabilitation research

committee reports 1998 gave an evaluation formula of the bond degradation as a function of

corrosion.

τcor = exp(-3.551×Wcr) ⋅ τnon-cor

The prediction with Japan Concrete Institute 1998 is in good agreement with S.18.C.0.08,

S.18.C.0.2, S.18.C.0.75 and S.18.C.0.85.However, our results are overestimated for 30 MPa

specimens.

Fig. 4.15 Comparison with fib model 2010 and JCI 1998

0.2 0.4 0.6 0.8 1

0.2

0.4

0.6

0.8

1

0Crack width (mm)

No

rm

ali

zed

lo

ad

18MPa specimens30MPa specimensModel Code 2010: upper limitModel Code 2010: lower limitJCI 1998

54

4.4.1.Bond degradation and surface crack width relationship

A simple formula for the bond strength with surface crack width as a variable is studied

similarly as Chap.3.

As can be seen from the results of regression analysis in Table 4.7, an exponential equation

provides the best fit (higher R2). Therefore, the following equation can be used for the

prediction of the maximum pull-out load:

P(Wcr) = P0 ⋅ e-0.-2.06Wcr

where P(Wcr): Pull-out strength in cracked concrete; P0: Pull-out strength of specimen

without crack; Wcr: Crack width.

Fig 4.16 shows the experimental result and the prediction model comparison. The ratios of

the experimental to calculated values are close to one, which indicates that the concrete crack

width can potentially be a good indicator to characterize bond strength degradation.

In Chap 3, a formula expressing the deterioration of the bond obtained with single splitting

type specimens has been proposed. In Fig.4.16, that formula is compared to the one obtained in

this chapter. It can be seen that the deterioration of the bond due to the induced cracks is more

severe in a side-splitting specimen than in single-splitting specimen. This result may be

explained by the fact that the number and position of induced crack can heavily affect the

deterioration of the bond.

Table 4.7 Regression analysis result

Analysis Equations R2

Linear -0.1.03Wcr+1 0.71

Logarithmic -0.24(Wcr) + 0.23 0.62

Exponential e-2.06 Wcr 0.82

Fig. 4.16 Result and prediction model

0.2 0.4 0.6 0.8 1

0.2

0.4

0.6

0.8

1

0Crack width (mm)

Norm

ali

zed

load

Present study 18MPaPresent study 30MPaPrediction model

e-2.06Wcr

55

Fig. 4.17 Proposed evaluation formula from test result

1 2 3

0.2

0.4

0.6

0.8

1

0Crack width (mm)

No

rm

ali

zed

lo

ad

Side-splittingSingle-splitting

56

4.5. Conclusions

In this chapter, the pull-out test is conducted on 16 concrete blocks with embedded D19 rebar

following different corrosion crack width simulated by 2 EAFPs. The test is carried out on a

side split type specimen. In summary, these results show that there was a significant exponential

correlation between pull-out strength and surface crack width. Also, a simple formula to predict

bond degradation is proposed by using the surface crack width as the main variable.

The prediction with Model Code 2010 is in good agreement only with S.30.C.0.25 and

S.30.C.0.30.

The prediction with Japan Concrete Institute 1998 is in good agreement with S.18.C.0.08,

S.18.C.0.2, S.18.C.0.75 and S.18.C.0.85.However, our results are overestimated for 30 MPa

specimens.

A comparison of the two results (single splitting and side splitting) reveals that the

deterioration of the bond due to the induced cracks was more severe in a side-splitting specimen

than in single-splitting specimen.

57

Chapter 5 Conclusions

The present study deals with the relationship between the bond strength of corroded steel

bars and the surface crack width. A total of 28 specimens were subjected to pull-out test. On the

basis of the experimental evidence and discussions presented in this thesis, the following

remarks can be drawn:

1) The expansion agent filled pipe is a promising method that allows focussing in the cracking itself. The tensile test shows that the strain in the axis, top and between the ribs

tended to increase over the time after the filling of the expansion agent. In addition, at the

beginning of the loading, with the increasing of the axial strain, the yield strength and

modulus of elasticity of the pipe filled with an expansion agent also increase. However,

the elongation at failure tended to decrease. The tensile strength was not affected.

2) The specimens were categorized in: “Single-split type” when the induced cracks are along the rebar and “Side split-type” when the induced cracks are located on the surroundings

and perpendicular to the rebar. All specimens experienced failure due to splitting. A group

of specimens failed by newly generated splitting crack despite existing induced crack and

other specimens failed by the opening of the induced crack.

3) There was a significant correlation between residual bond strength and induced crack width. This demonstrates that the surface crack width can potentially be a good indicator

to evaluate the bond strength degradation.

4) The decrease of the pull-out load is more severe in 18MPa specimens than in 30 MPa specimens.

5) The deterioration of the bond due to the induced cracks was more severe in a “Side-split type” than in “Single split-type”.

6) Empirical models, which relate the bond deterioration with the longitudinal cracks and the lateral cracks have been proposed. However, more research is necessary to investigate

the influence of other involved parameters (e.g. cover/diameter ratio, confinement, bar

profile) to develop a predictive model for general applicability.

58

DEDICACES

The work presented in this master thesis represents two years of study and research at the

University of Tsukuba.

The present work was carried out between April 2018 and March 2020 in the Kanakubo lab.

This was possible thanks to the scholarship received from the Japanese Government

Scholarship, funded by the Ministry of Education, Culture, Sports, Science and Technology

(MEXT). I am very appreciative for giving me this opportunity.

First of all, I would like to express my most sincere gratitude to my supervisor, Prof.

Kanakubo Toshiyuki. Throughout my journey in Japan, he has shared his vast knowledge and

experience, providing valuable discussion, advice and insights yet he has always given me the

freedom to conduct my own research, allowing me to evolve. Given his level of commitment

to my research, his understanding and his thorough review of my work, I cannot think of anyone

better to have as supervisor. Also, a big thanks to his wife, Kanakubo Noriko, for being a

valuable advisor during my life in Japan.

I want to convey my appreciation to my vice-supervisor, Prof. Yasojima Akira and Prof.

Shoji Gaku, for showing their interest in my work and taking the time to get involved and share

their valuable thoughts, comments and also having their door always open to share their vast

knowledge.

I would also like to thank all my colleagues, former and present, at Kanakubo laboratory for

creating such a nice working environment. Special thanks go to Aburano Togo and Kojima

Atsushi for their help in executing all the experimental work.

Finally, I want to direct my special thanks to my lovely family, my wife Aida, who has

been very understanding and patient during my study. Their love and affection have been

indispensable in this journey and none of this would have been possible without their

unconditional support. For all that and much more, thank you.

59

REFERENCES [1] R. Tefers, Cracking of concrete cover along rebars, Mag. Concr. Res. 31(106) (1976) 3-12

[2] fib Bulletin No. 10. (Bond of reinforcement in concrete 2000)

[3] Rodriguez et al.: Corrosion of reinforcing bars and service life of reinforced concrete

structures: corrosion and bond deterioration. Int. Conf. on Concrete across Borders, Odense,

Denmark, Vol. II, pp 315-326, 1994

[4] Auyeung et al: Bond-behavior of corroded reinforcement bars. ACI Structural Journal

97(2):214-220,2000

[5] Al-Sulaimani et al.: Influence of corrosion and cracking on bond behavior and strength of

reinforced concrete. Proceedings American Concrete Institute, Vol 87, No.2, Mar-Apr.

pp220-231 ,1990

[6] fib Model Code 2010, Vol.1. International Federation for Structural Concrete, Lausanne,

Switzerland.

[7] SYLL, A. S.，Kawamura, Y.，Kanakubo, T.: Simulation of Concrete Cracks due to Bar

Corrosion by Aluminum Pipe Filled with An Expansion Agent, Summaries of Technical Papers

of Annual Meeting, Architectural Institute of Japan, Structure IV, pp.55-56, 2018.9

[8] Yasojima, A., Kanakubo, T., Bond Splitting Strength of RC Members Based on Local Bond

Stress and Slip Behavior, 11th International Conference on Fracture, No.4486, 2005.

[9] Yang, Y., Nakamura, H., Miura, T., Yamamoto, Y., Effect of corrosion‐induced crack and

corroded rebar shape on bond behavior, Proceedings of the Japan Concrete Institute, Vol. 40,

No. 1, pp.927-932, 2019.

Documents

Study on Bond Degradation of Rebars in Cracked Concrete ...kanakubo/syllmas2020.pdf1 ABSTRACT This study aims to quantify the bond strength of reinforcing bars in cracked concrete